Pathological anxiety is commonly conceptualized as an exaggerated fear state in which hyper- excitability of fear circuits...is expressed as hypervigilance and increased behavioral responsivity to fearful stimuli (e.g., Rosen et al., 1998)-essentially an amplification of normal fear/defense circuit function. However, our previous research has shown that anxiety patients differ substantially in fear reactivity, within and across diagnoses, as measured by the magnitude of the probe startle reflex during fear imagery challenge (Cuthbert et al., 2003; Lang, McTeague & Cuthbert, 2005, 2007; Lang & McTeague, 2009). Thus, specific phobics show robust probe startle potentiation during fear imagery, while patients with more complex disorders (e.g, panic disorder, generalized anxiety disorder) characteristically show blunted reflex responses. Importantly, similar differences are also found within DSM diagnoses: Overall, fear potentiation decreases as the symptom picture is progressively more severe, comorbid, with higher scores on measures of negative affectivity-despite ratings of high, often much higher, fear during challenge (McTeague, Lang, et al., 2009, 2010, 2011). Considering that research both with animals and humans confirms that fear-circuit activation (amygdala mediated) potentiates the startle reflex, our research invites the novel hypothesis that the defense circuit is compromised/dysregulated in highly distressed anxiety patients. Thus, we examine reactivity to a fear imagery challenge in a broad sample of patients as they present for treatment at our health-center affiliated Fear and Anxiety Disorders Clinic (FADC). Functional MRI (fMRI) is used to asses fear/defense circuit function during fear imagery challenge, testing the hypothesis that alterations in normal defense circuit function mediates a dimension of fear reactivity that ranges from hyper-reactivity to hypo-reactivity in probe startle magnitude. A second broad aim is to test the hypothesis that increased circuit dysregulation and reflex blunting are reliably related to increased self-reports indexing a dimension of negative affectivity, as well as inversely related to treatment outcome success. The general research plan is therefore to develop biological classifiers for a dimension of pathology that cuts across anxiety/mood spectrum disorders. The approach is consistent with the NIMH Research Domain Criteria initiative (e.g, as described by Insel & Cuthbert 2009) implemented in RFA- MH-12-100.